24v-12v DC to DC Converter

[Formas]

Hey guys, im working on a 24 to 12volts converter, the original mosfets was RU6099R all was destroyed along with TL494 and IR2011.
Im unable to get the RU6099R and i was trying to use IRF4310 but they are heating up. The gate resistors are 47ohms, should i increase them?
square waves are fine both for the HS and LS, tried two 50N06 for test and they ran cool at idle, but when i loaded up the 8pcs of IRF4310 im having heating issues at idle.

 

[Perry Babin]

Typically, if the FETs in a power supply run hot at idle and the driver circuit is OK, you need to lower the gate resistor value.

 

[Formas]

Can 47ohms to 10ohms ok to be in the gate circuitry?

 

[Perry Babin]

For a production design, you'd want to find something that's a happy-medium. Too low and you make the drive circuit work much harder. Too high and the FETs run hot. You can try 10s but I'd try several values between 47 and 10 to see the highest value that will allow the FETs to run cool. You don't have to have all of the FETs in the circuit except for final testing if the drive circuit is something like an EF pair or a dedicated gate driver IC. You can try one in each bank. The final value may need to be slightly lower than the value required for only 1/bank due to inefficiencies in the drivers.

 

[rayma]

Can 47ohms to 10ohms ok to be in the gate circuitry?

Too low a value could cause shoot-through.

 

[Perry Babin]

I think that's only for class D or other circuits that can go to 50% duty cycle or higher.

 

[Formas]

This is the circuit, just as simple as it gets just an inductor, no toroidal transformer.
Heating up on the fets with all proper drive signals LS and HS.
I haven’t changed the gate resistors as yet (just gotten in to work).
The originals was RU6099R, I also changed both drive IC’s.
The IRFB4310 have an higher rating on volts and current but I don’t know about gate capacitance or underlying factors why it shouldn’t work.

 

[Perry Babin]

Are they using the 494 with pin 13 at 5v or 0v?

 

[Formas]

I would say 0, it’s connected to the ground circuit.

 

[Perry Babin]

I think lower is the way to go but you'll have to experiment. If the 2011 is driving the FETs directly, I'd stay as high as possible if you need to go lower to make the FETs run cooler.

 

[Formas]

I didn’t do the experiment, I just went for 22ohms it’s cooler (drawing 0.18amp at idle).
I only did the high side bank as they were the fastest to heat up, should I change also the low side gates? (Well obviously I’ve never seen different gates within a dual bank design).

 

[Perry Babin]

It may be OK only changing the resistors for the high side.

Is the driver IC remaining cool?

If you have a way to change the input voltage through the rated range for the converter, do so to see if the changing pulse width makes a difference in the heating.

 

[Formas]

I have adjustable bench PS, it’s just used on a 24volt vehicle to supply 12volts at (not exceeding 60amperes) for some headunit, drop down roof DVD player and a small amp.
But for your test I could swing it around upper or lower 24volts to see if it’s stable and thermally safe.
I’ve not the manual to know the parameters but I take it down to 18volts and it’s stable at 12.95volts minimal temp on the ic but one fet on the low side have a little temp but not hot to handle just a tad warm. Same to 26volts. There was adjusted duty cycle during the tests.